Disease burden, influencing factors and trend-based surveillance of needlestick injuries among healthcare workers: a seven-year longitudinal study and intervention evaluation at a tertiary hospital.

BACKGROUND: Needlestick injuries (NSIs) remain a leading occupational hazard among healthcare workers (HCWs), carrying substantial health and economic burdens. High-throughput procedural settings are particularly vulnerable, yet evidence on setting-specific interventions' effects and durability under healthcare system volatility remains limited. Existing surveillance systems for NSIs struggle to distinguish genuine risk escalation from clinical workload fluctuations. This study aims to characterize the epidemiological and economic burden of NSIs among HCWs, evaluate the effectiveness of safety interventions in the Infusion Center during post-pandemic clinical resurgence, and develop a dynamic NSIs surveillance system.

METHODS: A 7-year retrospective longitudinal study was conducted at a tertiary hospital across three phases: pre-, intra- and post-COVID-19 pandemic phases (P1, P2, and P3). The Infusion Center, which implemented a multi-component safety intervention bundle, served as the intervention group, with 38 other clinical departments as controls. NSIs incidence was modeled using a generalized linear mixed model (GLMM) with Poisson distribution, incorporating a department-level random intercept, a staff-size offset and hospital-wide clinical volume as a time-varying covariate. A dynamic early warning system was constructed using GLMM-derived, volume-adjusted upper control limits.

RESULTS: The Infusion Center exhibited a 22.43-fold higher NSIs risk than controls in P1 (IRR = 22.43, p < 0.001). Controls showed a 43% reduction in P2 (IRR = 0.57, p = 0.033) followed by a rebound in P3 (IRR = 1.06, p = 0.753). Conversely, the Infusion Center showed a sustained 66% reduction in P3 relative to P1 (IRR = 0.34, p = 0.0006), significantly diverging from the hospital-wide rebound trajectory (interaction IRR = 0.32, p = 0.005). The dynamic early warning system identified 25 alarms across controls during P3, while the Infusion Center triggered zero alarms. The intervention bundle was associated with 20 fewer NSIs cases in P3, yielding a direct medical cost avoidance of 12,776.68 RMB.

CONCLUSION: Systematic behavior-based interventions combining leadership accountability, competency training and continuous reinforcement are associated with a sustained reduction in NSIs risk that diverged from the post-pandemic rebound observed across control departments. The dynamic early warning system represents a proof-of-concept approach to volume-adjusted occupational safety surveillance, enabling targeted resource allocation to high-risk departments.